May 11, 2007 at 9:32 pm
· Filed under SoC2007, eclipse
My project was accepted on April 11. Currently the students are meant to be in the interim period “Students learn more about their project communities”. The coding starts on 28th of May. SoC page has a timeline up too.
About two weeks of the interim period I’ve spent in the woods. There was training going on to prepare for the grand training Kevadtorm 2007 starting tomorrow. I have not got out of the barracks for three weeks because of the current situation in Estonia. Luckily I’ve had some free time and I’m connected although to a very weak and unstable but still a WiFi access-point. What have I done turing this period:
- Compulsory paperwork to google
- Checking out the infrastructure (wiki, IRC, lists)
- Exchanging emails with my mentor and throwing different ideas around
- Played around with two JSR223 implementations for PHP
- Started learning Eclipse plugin architecture
- Different online articles (mostly for 3.1
)
- Learning from a plugin QuickREx
I have not spent too much time getting to know the community because of my situation of being offline so I’ve put my energy into research at this point. I’m going to be in the woods for the next 2 weeks. Completely offline. After that I’ll be back and for good. My service ends on 1st of June
. Stay tuned.
Permalink
May 11, 2007 at 8:09 pm
· Filed under SoC2007, eclipse
Although my application was accepted a long time ago I have not announced taking part in Google Summer of Code. I will use this blog to share my experience in this project and provide updates. The wiki page should become the second best place to get information. My mentor is AhtiK. The abstract of my application is
Framework to develop simple Eclipse plugin using PHP. Create a framework that could support developing simple Eclipse plugin using a PHP To Java bridge (Like Caucho’s Quercus and other JSR 223 engines). Develop and document a sample plugin in PHP, so that PHP developers could create simple plugins.
Permalink
May 11, 2007 at 7:54 pm
· Filed under SoC2007, eclipse
I usually have all the manuals/references/API docs bookmarked in a categorised folder. When I started out with a simple plugin for Eclipse I could not find the downloadable version of the API docs. I went through the eclipse.org’s downloads, documentation and resources sections. Then I checked the snapshots (after seeing its size I canceled it
) and then thought ok, CVS checkout. I tried to compile Eclipse from source about a year ago and I remember checkout taking a long time so I was stumped. Lastly I checked the Eclipse’s inapplication help section. Voila the API docs are bundled with the application itself.
As java software has a standard format for API docs that I’ve grown accustomed to I really can’t use the format imposed by the help system. A quick du -Sba | sort -nr | less (my previous search for html files proved the documentation not being in expanded format) in the installation folder of eclipse found ./plugins/org.eclipse.platform.doc.isv_3.2.2.R32x_v20070115.jar do be the largest single file. And in that file I found what I was looking for. Offline API docs in a folder api
Permalink
May 2, 2007 at 5:28 pm
· Filed under google, meme
I’m writing my bachelors thesis about packaging web software. One format i analyzed as prior work was Google Gadgets. As I’m still in the army and the situation here in Estonia is as it is I have lots of kinda free time with a really laggy WiFi internet connection.
Yesterday I thought I’ll give Google Gadgets a try from the developers point of view and started coding a small gadget. Google has great resources for this:
The problems I did ran into was testing my module @ personalised homepage. The problem is that the module gets cached and you have to change the URL the retest it. So the development cycle looks like this:
- develop the webpage
- prepare the xml file
- copy webpage to xml CDATA
- test (rename and test again)
Maybe there is a solution to this but I just don’t know about it. Anyways here is the gadget (original idea from duke’s page) I developed:
Permalink
May 1, 2007 at 12:57 pm
· Filed under chesspastebin, jsPGNViewer
Finally I had time to fix my darcs repository for chespastebin.com. Every now and then one my repos “break down”. Meaning that when pushing changes darcs will go crazy at the server end and run for ages with high CPU usage (well actually i don’t know if it would stop at one point, i just kill it). I did a manual merge of local changes and reinited the repo.
New features of chesspastebin.com:
- Include games in your pages with an iframe. Special minimalistic version has created for all games. It fits nicely into a frame, has no margins and looks nice. So just add a game and click the “Include the game as a frame in your page” at the game page and follow the instructions. Here is a sample of a no margin page.
- Commenting games. You can add comments to games just like comments to blog posts. When I wrote it couple of months ago i thought it would be cool. At the moment i still have it disabled. I’ll enable it when there is enough interest or i again find the need.
- The number of games is available under the Random game link.
- Updated jsPgnViewer to version 0.3.7 (well actually to trunk:)).
I have been looking for JavaScript code minimizers and obfuscators. I ran into JSMin and Dean Edwards Packer. To make them work I had to change a lot of code. I like to write JavaScript without using the line terminating semicolons. Minimizers and obfuscators obviously don’t like
. I went over the source code and I will release a lighter version of jsPgnViewer soon.
With JSMin I was able to achieve 56K -> 34K change. The Packer was able to minimize 56K into 22K. Thats a lot!
Head off to chesspastebin.com and check out the new features.
Permalink